一般所有存活分析中都是假設事件時間是易感受性,也就是假設每事件必然發生。 在臨床試驗中,這樣的假設對長期存活資料並不適當。在本計畫中,我們將針對事件未 必發生(不易感受性)及存在變係數問題研究探討,更進一步建立線性轉換治癒模型。我 們提出平滑曲面下的最大概似估計量,並證明其一致性與收斂性。期模擬研究結果能與 理論相對應。 Varying Coefficient Transformation Cure Models with Censored Data The phenomenon of disease may be completely eliminated, that is a significant fraction of patients can be cured; also, the phenomenon of covariates may be time-varying. The method considered in this present paper applies to covariates measured to time-varying covariates is also possible as long as the covariates are ancillary or external with transformation cure models. A maximum likelihood method with spline smoothing is proposed and the estimators, under some regularity conditions, are proved to be consistent and asymptotically normal. A simulation study is presented to show that the proposed method performs well with finite sample and is easy to use in practice.